Download A PEST-like Sequence in the N-Terminal Cytoplasmic Domain of

4518
Biochemistry 2000, 39, 4518-4526
A PEST-like Sequence in the N-Terminal Cytoplasmic Domain of Saccharomyces
Maltose Permease Is Required for Glucose-Induced Proteolysis and Rapid
Inactivation of Transport Activity†
Igor Medintz,‡ Xin Wang, Thomas Hradek, and Corinne A. Michels*
Biology Department, Queens College and the Graduate School of the City UniVersity of New York, 65-30 Kissena BouleVard,
Flushing, New York 11367
ReceiVed October 22, 1999; ReVised Manuscript ReceiVed February 7, 2000
ABSTRACT: Maltose permease is required for maltose transport into Saccharomyces cells. Glucose addition
to maltose-fermenting cells causes selective delivery of this integral plasma membrane protein to the
yeast vacuole via endocytosis for degradation by resident proteases. This glucose-induced degradation is
independent of the proteasome but requires ubiquitin and certain ubiquitin conjugating enzymes. We
used mutation analysis to identify target sequences in Mal61/HA maltose permease involved in its selective
glucose-induced degradation. A nonsense mutation was introduced at codon 581, creating a truncated
functional maltose permease. Additional missense mutations were introduced into the mal61/HA-581NS
allele, altering potential phosphorylation and ubiquitination sites. No significant effect was seen on the
rate of glucose-induced degradation of these mutant proteins. Deletion mutations were constructed, removing
residues 2-30, 31-60, 61-90, and 49-78 of the N-terminal cytoplasmic domain, as well as a missense
mutation of a dileucine motif. Results indicate that the proline-, glutamate-, aspartate-, serine-, and threoninerich (PEST) sequence found in the N-terminal cytoplasmic domain, particularly residues 49-78, is required
for glucose-induced degradation of Mal61/HAp and for the rapid glucose-induced inactivation of maltose
transport activity. The decreased rate of glucose-induced degradation correlates with a decrease in the
level of glucose-induced ubiquitination of the ∆PEST mutant permease. In addition, newly synthesized
mutant permease proteins lacking residues 49-78 or carrying an alteration in the dileucine motif, residues
69 and 70, are resistant to glucose-induced inactivation of maltose transport activity. This N-terminal
PEST-like sequence is the target of both the Rgt2p-dependent and the Glc7p-Reg1p-dependent glucose
signaling pathways.
Maltose-fermenting Saccharomyces cells require maltose
permease to transport maltose across the plasma membrane
and for induction of MAL gene expression (1, 2). Addition
of glucose to a maltose-fermenting culture causes a biphasic
loss of transport activity, characterized by an initial very rapid
inactivation of maltose transport activity followed by a slower
loss in transport activity that correlates with the degradation
of maltose permease protein (3). This degradation is dependent on endocytosis, vesicle sorting, and the vacuolar
proteolysis enzymes and is independent of the proteasome
(3). As has been shown for a number of yeast and
mammalian plasma membrane proteins, we found that
ubiquitin and specific components of the ubiquitin conjugating system play an important role in the glucose-induced
† This work was supported by grants to C.A.M. from the National
Institute of General Medical Sciences (GM28216 and GM49280). T.H.
received an undergraduate research stipend from a Howard Hughes
Biomedical Institute grant to Queens College. This work was carried
out in partial fulfillment of the requirements for the Ph.D. degree from
the Graduate School of the City University of New York (I.M. and
X.W.).
* Corresponding author. Telephone: (718)-997-3410. Fax: (718)997-3431. E-mail: [email protected].
‡ Current address: Department of Chemistry, University of California
at Berkeley, Berkeley, CA 94720.
degradation of maltose permease (reviewed in refs 4 and 5).
In this report, we use mutation analysis of the N- and
C-terminal cytoplasmic domains of maltose permease to
identify sequences required for the glucose-induced inactivation and proteolysis of this integral plasma membrane protein.
Mutation analysis was used to define the ubiquitination
and endocytosis targeting signals in several yeast membrane
proteins. These include the SINNDAKSS motif in the
C-terminal cytoplasmic domain of the Ste2 R-factor receptor
(6, 7). Deletion analysis demonstrated that the middlespanning linker region of the ABC transporter, Ste6p,
contains a signal that mediates ubiquitination and rapid
turnover of this protein (8, 9). PEST sequences, regions rich
in proline, glutamate, aspartate, serine, and threonine, are
found in many short-lived proteins. Phosphorylation of serine
and threonine residues in PEST sequences is suggested to
be associated with regulated degradation, as has been
demonstrated for a putative PEST-like sequence in Fur4p,
the Saccharomyces uracil permease (10, 11). Other internalization signals have been described in yeast membrane
spanning proteins that do not appear to be associated with
regulated phosphorylation including the dileucine motif in
the general amino acid permease Gap1 and the NPF signal
of the a-factor receptor Ste3 (12, 13).
10.1021/bi992455a CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/22/2000
Mutation of Maltose Permease
Biochemistry, Vol. 39, No. 15, 2000 4519
Table 1: List of Oligonucleotides Used for in Vitro Mutagenesis of MAL61/HAa
MAL61/HA allele
C-terminal mutations
mal61/HA-581NS
mal61/HA-581NS(P577A)
mal61/HA-581NS(P566A)
mal61/HA-581NS(S572A,
T573A)
mal61/HA-581NS(K569A,
K571A,K574A)
N-terminal mutations
mal61/HA-(∆2-30)
mal61/HA-(∆31-60)
mal61/HA-(∆61-90)
mal61/HA-(∆49-78)
mal61/HA-(∆49-60)
mal61/HA-(∆61-78)
mal61/HA-(L69A,L70A)
oligonucleotide sequence (5′-3′)
annealing site
GCAGCTGCTGCTTTTCAAGCTGCAAAAGG
AGCTGCAAAAGCGTCGACTTTAGT
CTTTCTTGCTGCAACACCAAG
GCAAAAGGGTCGACTTTAGCCGCCTTGAACTTGAACTTTCTTGC
1757-1729
1740-1717
1707-1687
1736-1699
AGGGTCGACTCTAGTCGATCTGAATCTTCTTGCTGGAACACC
1731-1699
TTGCTCCTCCATGTCTATCGAGTTCAAAGGACCACCCAAGCTAG
CGTAATCTGG
GAGAAGGTCGGGGACTTCTTCATTATTATCTTCGGTAGCGTTCA
CGCCATTCTCGATCTC
CCAAGCAGCAGCTTTTGGATATGTCTTCAAAGCTGTGTTTGGTAT
TAGTGAACCTGGACCGTACTC
GAGTGGCATTCCCCTCTCACTTTCATCTGCCTCGGAAAGATCAA
AATCACTTTTCTTACCTTGCTCCTCC
CATCGAGAAGGTCGGGGACTTCTTCATTATTATCGGAAAGATCA
AAATCACTTTTCTTACCTTGCTCCTCC
GAGTGGCATTCCCCTCTCACTTTCATCTGCCTCGTTTGGTATTAG
TGAACCTGGACCGTACTC
GGCGTCCTGCATAGCTTCATCGGCAGCGTCGGGGACTTCTTC
117-91, HA tag
210-181, 90-61
306-271, 180-151
267-235, 145-110
214-183, 145-110
276-235, 180-151
231-191
The nucleotide numbers are based on the sequence of MAL61 (2) with base +1 as the first base in the MAL61 open reading frame and do not
include the HA tag (described in ref 3). The 5′ 26 bases of the oligonucleotide used to create mal61/HA-(∆2-30) are complementary to this HA-tag
sequence. Underlines indicate the mutant alterations.
a
Mal61/HA1 maltose permease is found in both phosphorylated and dephosphorylated forms in maltose-fermenting
cells, and our previous studies indicate that the level of
phosphorylation increases in the presence of glucose (3). The
N-terminal, C-terminal, and central cytoplasmic domains of
Mal61/HA maltose permease are rich in serine and threonine
and contain several potential phosphorylation sites (see
Figure 1) (2). A PEST region spans residues 48-79 of the
N-terminal cytoplasmic domain (2). Brondijk et al. (14)
investigated the role of several of these sites located in the
central cytoplasmic domain of Mal61p and found that serine295, located immediately C-terminal to TMD 6, is essential
for glucose-induced proteolysis of maltose permease. Interestingly, the rate of glucose-induced inactivation of maltose
transport activity is only slightly decreased in this mutant,
indicating that permease internalization is occurring at near
normal rates but delivery to the vacuole or proteolysis is
compromised. We undertook to define sites in Mal61/HAp
involved in the initial steps of internalization. Our results
suggest that the N-terminal PEST region mediates glucoseinduced internalization and thereby controls the initiating
events for the proteolysis of maltose permease. Deletion of
the PEST sequence decreases the level of glucose-induced
ubiquitination of Mal61/HA maltose permease. An additional
glucose effect on the secretion and/or localization of maltose
permease was identified that also requires residues in the
PEST region, particularly the dileucine motif. The C-terminal
cytoplasmic domain does not appear to be involved in either
of these functions.
MATERIALS AND METHODS
Strain Construction. Strain CMY1001 (genotype MATa
MAL61/HA MAL12 MAL13 leu2 ura3-52 lys2-801 ade2-101
trp1-∆63 his3-∆200) contains the MAL61 hemagglutintagged gene construct as described in Medintz et al. (3). It
1
Abbreviation: HA, hemagglutinin epitope.
contains a single MAL1 locus, and the maltose permease gene
at this MAL1 locus is replaced by the epitope-tagged MAL61/
HA allele derived from the MAL6 maltose permease gene
MAL61. A mal61/HA null derivative of CMY1001 was
constructed by PCR-based gene disruption as follows (15).
The upstream (5′) primer consists of 45 bases of sequence
complementary to the upstream region of MAL61/HA,
followed by 19 bases of sequence complementary to the
HIS3 gene 5′ sequence. Its sequence is 5′GTACTCAGCATATAAAGAGACACAATATACTCCATACTTGTTGTGAGTGGGCTTGGTGAGCGCTAGGAG3′. The
downstream (3′) primer contains 45 bases of sequence
complementary to sequence 3′ of MAL61/HA, followed
by 21 bases 3′ to the HIS3 ORF. The sequence of the
downstream primer is 5′CACAACAGATGGGGTGCTTCGCCCTTCATCTACCACAGAAGTTTCCAAATCCACACCGCATAGATCCGTCG3′. These primers were used to amplify the HIS3 gene from plasmid
pRS413 (18), and this amplified product was used for a onestep gene replacement of the MAL61/HA sequence present
at MAL1 of CMY1001, creating strain CMY1050. Gene
replacement was confirmed by Southern analysis. The same
process was used to delete the MAL61/HA gene in the end3ts strain CMY1004 (3) to create strain CMY1051 and in the
rgt2∆::KANR strain CMY1009 (17) to create strain CMY1052.
Plasmid Construction and Mutagenesis. MAL61/HA was
constructed from MAL61, encoding maltose permease at the
MAL6 locus, by inserting a 12-codon sequence containing
the HA epitope tag at the N-terminus of the Mal61 protein
(3). MAL61/HA was cloned into the vector pUN30 (16),
yielding plasmid pMAL61/HA. This plasmid was used as a
template for in vitro mutagenesis using the Bio-Rad MutaGene kit (Bio-Rad, Richland, CA) (19). The mutagenic
primers used are listed in Table 1. Alterations in MAL61/
HA were confirmed by sequencing. All mutant alleles were
constructed with the full-length MAL61/HA gene as a
template except for certain C-terminal cytoplasmic domain
4520 Biochemistry, Vol. 39, No. 15, 2000
Medintz et al.
FIGURE 1: Sequence of mutations in the N-terminal and C-terminal cytoplasmic domains of Mal61/HA maltose permease. The rectangular
shaded areas represent transmembrane domains. The position of the hemagglutinin epitope tag at the N-terminus is indicated as a circle
with the letters HA. Arrows point to altered residues. The residue numbers relate to the sequence of the wild-type Mal61 protein reported
by Cheng and Michels (2) and do not include the HA tag.
mutations that used the 581-stop mutant allele of MAL61/
HA as a template.
Plasmid pRGT2-1, provided by S. Ozcan and M. Johnston,
contains the RGT2-1 allele, a constitutive signaling allele of
the high-glucose sensor (20, 21).
InactiVation Assay Protocol. The inactivation assay protocol is described in detail in Medintz et al. (3). Briefly,
strains are grown at 30 °C to very early log phase (OD600 of
0.1-0.3) in maltose-containing selective medium. Cells are
harvested and transferred to nitrogen starvation medium
(yeast nitrogen base without amino acids and ammonium
sulfate) plus 2% glucose (or 3% ethanol as indicated). Cell
samples are collected at the indicated times over a 3 h period,
and for each sample, maltose transport activity is determined
and total cell extracts are prepared for Western analysis.
Growth dilution is calculated as the OD600 at time 0 divided
by OD600 at time x. All inactivation assays were done at least
in duplicate cultures of independent transformants. Transport
assays of each sample were done in duplicate, and the
standard error was less than 20%.
Western Blotting. Western blotting analysis was carried
out as described previously (3). The Mal61/HA protein in
the extracts is detected using anti-HA-specific antibody and
the ECL Western blotting kit (Amersham) or the Vistra-ECF
kit (Amersham) and the Storm 860 image analyzer (Molecular Dynamics). The relative protein levels in each sample
were determined using the Storm 860 image capture software
by densitometric comparison to the zero time sample.
Western analysis was done in duplicate on extracts prepared
from duplicate experiments carried out with at least two
independent transformants. To determine the half-life of the
mutant proteins, the data from the Western analysis of several
experiments were plotted and linear least-squares analysis
was used to determine the line of best fit and the time point
at which 50% of the initial Mal61/HA protein remained.
Sugar Transport Assays. Maltose transport was measured
as the rate of uptake of 1 mM [14C]maltose as described in
Cheng and Michels (2) and Medintz et al. (3). Assays were
done in duplicate on two to three transformants.
RESULTS
Mutation Analysis of the Mal61/HA C-Terminal Cytoplasmic Domain. The C-terminal cytoplasmic domain of Mal61
maltose permease contains several potential phosphorylation
Mutation of Maltose Permease
Biochemistry, Vol. 39, No. 15, 2000 4521
Table 2: Phenotype of Mutations in the C-Terminal Cytoplasmic
Domain of Mal61/HA Maltose Permeasea
MAL61/HA allele
MAL61/HA
mal61/HA-581NS
mal61/HA-581NS
(P577A)
mal61/HA-581NS
(P566A)
mal61/HA-581NS
(S572A,
T573A)
mal61/HA-581NS
(K569A,K571A,
K574A)
vector control
relative
maltose
half-life
level of
transport of maltose
maltose Mal61/HA
activity
permease
fermen- protein (% (nM min-1 protein in
tation of wild type)
mg-1)
glucose (h)
+
+
+
100
91
83
5.96
1.71
1.82
1.7
2
0.5
+
72
1.30
2
+
67
0.71
2.4
+
99
0.95
1.2
-
-
0.06
ND
a
MAL61/HA and the mutant alleles are carried on the CEN vector
pUN30 [Elledge and Davis (18)] and were introduced into the maltose
permease null strain CMY1050 (mal61∆::HIS3) for analysis of
phenotype. The Mal61/HA maltose permease relative protein levels
were determined by comparison of Western blots using total cell extracts
from cells grown in selective medium lacking tryptophan plus 2%
maltose. Maltose transport activity was determined in at least duplicate
transformants grown in selective medium lacking tryptophan plus 2%
maltose. The half-life of maltose permease was determined as described
in Materials and Methods from the results of glucose-induced inactivation assays performed in duplicate on at least duplicate transformants.
ND ) not determined.
sites and nearby lysine residues (Figure 1). Using sitedirected mutagenesis a nonsense mutation was introduced
at codon 581 truncating the C-terminal cytoplasmic domain
to approximately 36 residues (see Figure 1A). This mal61/
HA-581NS allele, when introduced into strain CMY1050
carrying a deletion of the chromosomal maltose permease
gene, complements the maltose-nonfermenting phenotype.
Transformants carrying mal61/HA-581NS are maltose inducible, ferment maltose, and express approximately normal
levels of mutant maltose permease protein (determined by
Western blot) and maltase activity, but maltose transport
activity is approximately 25-35% that of transformants
expressing MAL61/HA (Table 2). The glucose-induced halflife of mal61/HA-581NS permease is not significantly
different from that of the wild-type Mal61/HAp, 2 h vs 1.7
h.
Additional missense mutations were introduced into the
mal61/HA-581NS allele, altering residues from 566 to 574
that might affect secondary structure, sites of phosphorylation, or ubiquitination of the truncated C-terminal cytoplasmic domain. These include mal61/HA-581NS(P577A), mal61/
HA-581NS(P566A), mal61/HA-581NS(S572A,T573A), and
mal61/HA-581NS(K569A,K571A,K574A). When introduced
into CMY1050, all of the mutant maltose permease proteins
are expressed at approximately normal levels and transformants are capable of fermentative growth on maltosecontaining media (see Table 2). Maltose transport activity
of strains expressing these mutant permeases range from 12%
of wild-type levels for mal61/HA-581NS(S572A,T573A) to
30% for mal61/HA-581NS(P577A). The rate of glucoseinduced proteolysis of these mutant permeases is not
significantly changed, and in some cases the rate is increased.
The mutant permease mal61/HA-581NS(P577A) exhibits a
Table 3: Phenotype of Mutations in the N-Terminal Cytoplasmic
Domain of Mal61/HA Maltose Permeasea
MAL61/HA allele
MAL61/HA
mal61/HA(∆2-30)
mal61/HA(∆31-60)
mal61/HA(∆61-90)
mal61/HA(∆49-78)
mal61/HA(∆49-60)
mal61/HA(∆61-78)
mal61/HA(L69A,L70A)
vector control
a
maltose
relative
half-life of
transport
level of
maltose
maltose activity
Mal61/HA permease
fermen- (nM min-1 protein (% protein in
tation
mg-1) of wild type) glucose (h)
+
+
+
+
+
+
+
-
5.96
0.71
2.07
0.07
0.43
0.83
0.53
0.53
0.06
100
85
76
34
71
126
81
111
ND
1.7
3.5
>4
>4
>4
3.4
>4
1.5
ND
See Table 2 for details. ND ) not determined.
glucose-induced half-life of 0.5 h, approximately 3-fold faster
than that of the wild-type protein.
Mutation Analysis of the Mal61/HA Permease N-Terminal
Cytoplasmic Domain. Mal61/HAp contains an N-terminal
cytoplasmic PEST sequence that spans residues 49-78, and
several additional serine residues flank this site (2). Located
within the PEST sequence is a dileucine motif at residues
69 and 70. To investigate residues critical for glucoseinduced inactivation, a series of deletion mutations of this
domain along with specific point mutations were constructed.
These include deletions of residues 2-30, 31-60, 61-90,
49-78, 49-60, and 61-78 and point mutations of the
dileucine motif at residues 69 and 70 to alanine (see Figure
1). Plasmid-borne mutant alleles were transformed into strain
CMY1050 and the phenotypes determined (see Table 3).
Only mal61/HA(∆61-90) is unable to complement the
nonfermenting phenotype of CMY1050. Transformants carrying the other mutant alleles all expressed reduced levels
of maltose transport activity, ranging from 7% to 35% of
wild type. The relative level of mutant Mal61/HA protein
expressed in these cells was determined by Western blots
of total cell extracts prepared from transformants grown in
synthetic medium with 2% glycerol/3% lactate plus 2%
maltose (Table 3). These do not vary more than 25-30%
from wild-type expression levels with the exception of
Mal61/HA(∆61-90)p. While Mal61/HA(∆61-90) mutant
permease lacks significant maltose transport activity, expression of the mutant protein is maltose inducible and about
one-third the level of wild type.
Figure 2 presents the results of glucose-induced inactivation assays carried out on these N-terminal mutants. Transformants expressing mutant permeases lacking all or a
portion of the PEST sequence exhibit dramatically reduced
rates of glucose-induced degradation. In transformants
expressing mutant permeases with deletions of residues 4978, 49-60, and 61-78, glucose stimulates an immediate
increase in maltose transport activity, from 3- to 9-fold.
Variation in the level of increase is seen among different
transformants but appears to be constant in a particular
transformant (data not shown). Lineweaver-Burk kinetic
analysis of mal61/HA(∆49-78) maltose transport at time 0
and 1 h after the addition of glucose found that the increased
transport activity results from an increase in Vmax of the
transporter and not from a change in Km (data not shown).
During the 3 h time course of inactivation experiments, the
4522 Biochemistry, Vol. 39, No. 15, 2000
Medintz et al.
FIGURE 2: Glucose-induced inactivation of maltose permease mutant alleles with alterations in the N-terminal cytoplasmic domain. CMY1050
transformants carrying either plasmid pMAL61/HA, pMAL61/HA(∆2-30), pMAL61/HA(∆31-60), pMAL61/HA(∆49-78), pMAL61/
HA(∆49-60), pMAL61/HA(∆61-78), pMAL61/HA(∆60-90), or pMAL61/HA(L69A,L70A) were grown on selective medium plus 2%
maltose, harvested, and transferred to nitrogen starvation medium plus 2% glucose. At the indicated times the OD600 was determined, and
aliquots of the culture were removed for maltose transport assays and the preparation of total protein extracts for Western analysis of
Mal61/HA protein levels, as described in Materials and Methods. The Vistra-ECF Western blotting visualization system (Amersham) and
the Molecular Dynamics storm image analyzer were used to capture the Western image and quantitate the relative density of each time
point. Representative Western blots are shown for strains carrying pMAL61/HA and pMAL61/HA(∆49-78). The quantitation data used
in the graphs were obtained from the average of at least two experiments, each run on duplicate gels, and scanned twice. The relative levels
of Mal61/HA protein (b) and maltose transport activity (O) compared to the zero time sample are plotted along with the growth dilution
(9, dotted line). Growth dilution represents the growth of the culture during the course of the experiment and is calculated as the OD600 at
time 0 divided by the OD600 time x.
maltose transport activity remains at this elevated level,
indicating that internalization of these mutant permeases is
blocked. Thus, loss of residues 49-78 appears to provide
resistance to internalization.
The strain expressing the mal61/HA(L69A,L70A) permease reproducibly exhibits an approximately 50% increase
in maltose transport activity about 1 h after glucose addition,
but in contrast to the ∆PEST mutations, transport activity
drops to 20% of initial levels by the third hour (Figure 2).
Glucose-induced degradation of this dileucine motif mutant
permease protein appears to be normal. Mutation of the
dileucine motif provides resistance to the rapid glucoseinduced inactivation of maltose transport activity but not
internalization and proteolysis.
Glucose-Induced Ubiquitination of the ∆PEST Mutant
Maltose Permease. Previous studies demonstrated that
glucose stimulates ubiquitination of maltose permease and
that this ubiquitination is involved in the glucose-induced
degradation of maltose permease (5). If the PEST sequence
is required for ubiquitin conjugation, we should expect
significantly reduced rates of ubiquitination of the mutant
protein. Ubiquitination of maltose permease is most effectively demonstrated in an endocytosis-deficient strain
expressing abundant levels of a c-myc-tagged ubiquitin allele
(5, 22). Strain CMY1052 contains a temperature-sensitive
end3 allele and is deficient in endocytosis at the nonpermissive temperature of 37 °C (3, 23). Strain CMY1052 was
transformed with plasmid pMyc-Ub carrying the c-myctagged ubiquitin gene expressed from the CUP1 copperinducible promoter (24) and either MAL61/HA or mal61/
HA(∆49-78). Transformants were grown under conditions
optimized for demonstrating ubiquitinated Mal61/HA maltose
Mutation of Maltose Permease
FIGURE 3: Relative glucose-induced ubiquitination of wild-type and
∆PEST mutant Mal61/HA maltose permease. Strain CMY1052 was
transformed with plasmid YEp105, pCUP1-mycUb [Medintz et al.
(5)] carrying a c-myc-tagged ubiquitin allele expressed from the
copper-inducible CUP1 promoter and with the indicated MAL61/
HA allele. Cells were grown in selective synthetic media with 2%
maltose to early log phase and incubated in 0.1 mM CuSO4 for 3
h at 30 °C. Cultures were transferred to 37 °C for 1 h, and then
2% glucose was added to the growth media for an additional 30
min prior to harvesting. Samples were prepared for Western blotting
as described in Materials and Methods and Figure 2. The arrow
indicates the position of the monoubiquitinated Mal61/HAp. The
positions of the molecular weight markers are indicated.
permease (5). Samples containing equal amounts of maltose
permease protein were analyzed by Western blot. The
predicted molecular mass of Mal61/HA protein is approximately 69 kDa, but its apparent molecular mass is less,
probably due to the hydrophobicity of the protein. As seen
in Figure 3, a slower mobility species of Mal61/HAp is
visible in the control lane, which our previous studies
demonstrated to be a monoubiquitinated species of Mal61/
HA maltose permease (3). Only a single band is seen in the
strain expressing Mal61/HA(∆49-78) permease with an
apparent size consistent with the loss of 30 residues. Little
or no ubiquitinated Mal61/HA(∆49-78) permease is evident
at the appropriate molecular mass. The reduced rate of
glucose-induced proteolysis of this mutant protein correlates
with a reduced level of ubiquitination.
InactiVation of Mal61/HA Maltose Permease Mutants in
Pathway 1 and Pathway 2 Strains. Jiang et al. (17) identified
two glucose sensing/signaling pathways in Saccharomyces
that are used to monitor high-glucose levels and stimulate
glucose-induced degradation of maltose permease. To determine whether N-terminal residues 49-78 are the target
of glucose signaling pathways 1 or 2 or both, strains were
constructed that carry a deletion of the chromosomal maltose
permease gene and alterations in these signaling pathways.
These were transformed with either plasmid-borne MAL61/
HA or mal61/HA(∆49-78) and underwent glucose inactivation assays. The panels to the left in Figure 4 show the results
from strain CMY1050, genotype RGT2, in which both
pathways 1 and 2 are functional. The results in the middle
panels were obtained with strain CMY1051, genotype rgt2∆,
which is isogenic to CMY1050 except that the RGT2 ORF
is replaced with the KANR gene (15, 17). Since the gene
encoding this high-glucose sensor of pathway 1 has been
deleted, strain CMY1051 is deficient in pathway 1 signaling
Biochemistry, Vol. 39, No. 15, 2000 4523
and only has a functional pathway 2. Introduction of the
dominant constitutive RGT2-1 allele carried by plasmid
pRGT2-1 into strain CMY1050 activates pathway 1 in the
absence of glucose without activating pathway 2 (17). Results
from strain CMY1050[pRGT2-1] are shown in the right
column of panels in Figure 4. The results for each of these
strains expressing wild-type Mal61/HA maltose permease
are shown in the top row, and those for strains expressing
Mal61/HA(∆49-78)p are shown in the bottom row.
The top row of panels confirms the results of Jiang et al.
(17), showing that pathway 1 is responsible for most of the
observed rate of maltose permease proteolysis but not the
rapid loss in transport activity. Despite the fact that maltose
permease protein levels are decreasing in the RGT2-1 strain,
maltose transport activity undergoes an approximately 2550% transient increase, also found by Jiang et al. (17).
Pathway 2 is solely responsible for the rapid glucose-induced
inactivation of maltose transport activity and has only modest
effects on the proteolysis of maltose permease (17). In strain
CMY1050 (RGT2) where both pathways 1 and 2 are active,
little or no proteolysis of Mal61/HA(∆49-78) is stimulated
by glucose. Also, in the constitutive pathway 1 strain (RGT21) no turnover of Mal61/HA(∆49-78) permease is detected.
The effect of the ∆49-78 deletion is not as obvious in the
rgt2∆ strain. Glucose induces a rapid loss in transport activity
in wild-type permease, more rapid than can be explained by
loss of permease protein, and this is dependent on pathway
2 signaling (3, 17). In the strain expressing only signaling
pathway 2 (middle panels of Figure 4), one observes a
significant decrease in this rapid inactivation in the strain
expressing the ∆49-78 mutant permease. This suggests that
the pathway 2 target sequence is also located in residues
49-78.
DISCUSSION
The predicted topology of Mal61p suggests that the
N-terminal 109 residues, the C-terminal 65 residues, and the
central region of 72 residues are each positioned on the
cytoplasmic side of the plasma membrane and thus represent
potential target sites for stimulating glucose-induced inactivation and proteolysis (2). The N-terminal cytoplasmic
domain of the yeast uracil permease, Fur4p, contains a PESTlike sequence shown to be essential for turnover (11).
Internalization of Fur4p was reduced when some of the serine
residues in the PEST-like sequence were converted to
alanines and severely impaired when all serines in this region
were mutated or when this region was deleted (11). A PESTlike sequence, proposed to be a site of regulated phosphorylation and ubiquitination, is located in the N-terminal
cytoplasmic domain of Mal61p (see Figure 1) (2). Other
potential phosphorylation sites are found in each of the
cytoplasmic domains, and lysine residues capable of serving
as ubiquitination sites are positioned nearby these sites (14).
A dileucine motif is found at residues 69 and 70 of Mal61p.
A dileucine motif of GLUT4, the human insulin-responsive
glucose transporter, was found to be required for internalization and also is proposed to play a similar role in the yeast
Gap1 general amino acid permease and the eukaryotic β2adrenergic receptor (12, 25). Serine 295, located just after
transmembrane domain 6, is essential for glucose-induced
proteolysis of Mal61 maltose permease (14). Alteration of
this residue to alanine had only a modest impact on the
4524 Biochemistry, Vol. 39, No. 15, 2000
Medintz et al.
FIGURE 4: Glucose-induced inactivation of pathway 1 and pathway 2 deficient strains expressing mutant Mal61/HA maltose permease.
Results of inactivation assays carried out in strain CMY1050 (RGT2), CMY1051 (rgt2∆), or CMY1050[pRGT2-1] expressing either MAL61/
HA or mal61/HA(∆49-78) are presented. Glucose was used to stimulate inactivation in the results shown in the left and middle panels
(YNSG indicates inactivation media containing 2% glucose). The right column of panels presents inactivation assays carried out in 2%
ethanol as the carbon source and represents effects of constitutive RGT2-1 signaling (YNSE indicates inactivation media containing 2%
ethanol). Samples were taken at the indicated times, and the growth dilution (9), maltose transport activity (O), and relative Mal61/HA
protein levels (b) were determined as described in Materials and Methods and Figure 2.
glucose-stimulated loss of maltose transport activity (14),
suggesting that S295 is not involved in the internalization
of maltose permease but in a later event required for
proteolysis. Glucose stimulates increased phosphorylation
and ubiquitination of maltose permease, which are required
for internalization and degradation (3, 5). We undertook a
mutation analysis of sequences in the N- and C-terminal
cytoplasmic domains of Mal61/HA to identify sites critical
to the early steps of glucose-induced inactivation and to
identify pathways by which glucose marks maltose permease
for internalization and degradation.
The cytoplasmic C-terminal regions of several yeast
plasma membrane proteins contain sequences involved in
their internalization, including Ste2 R-factor receptor, Gap1
general amino acid permease, and Ste3 a-factor receptor (6,
7, 13, 26). Alterations in the C-terminal cytoplasmic domain
of Mal61 permease have variable but relatively minor effects
on the rate of glucose-induced proteolysis (see Table 2).
Mutation mal61/HA(NS581) produces a truncated protein
with two proline residues, one serine and one threonine
residue, and three lysine residues found in the last 15
C-terminal residues. This truncated protein is still sensitive
to glucose-induced inactivation and proteolysis (Table 2).
Mutations altering the proline residues at 566 or 577 to
alanine, the potential protein kinase C phosphorylation sites
at S572 and T573 to alanine, and all of the lysine residues
at 569, 571, and 574 to alanine have little effect on the
glucose-induced proteolysis of these truncated mutant permeases. These results suggest that residues in the C-terminal
domain do not play an important role in glucose-induced
proteolysis.
The PEST sequence of Mal61 maltose permease is an
important targeting sequence for the degradation of this
protein. Mutant Mal61 proteins lacking this sequence, or
portions thereof, exhibit significantly reduced rates of
glucose-induced proteolysis (Table 3 and Figure 2). While
we have not demonstrated the site(s) of ubiquitin conjugation,
we do show a dramatic reduction in the relative amount of
the ubiquitin-conjugated form of the Mal61(∆49-78) protein
(Figure 3). Thus, residues 49-78 are required for efficient
glucose-induced ubiquitination and proteolysis of maltose
permease. Several lysine residues lie in and around the PEST
sequence and could serve as ubiquitin conjugation sites. Since
some effect is seen in deletion ∆2-30, it is likely that
Mutation of Maltose Permease
additional residues outside of the PEST(49-78) are also part
of the target site.
Two glucose signaling pathways regulate glucose-induced
inactivation of maltose permease (17). Pathway 1 utilizes
Rgt2p, a sensor of high concentrations of extracellular
glucose that also regulates HXT1 expression encoding a lowaffinity glucose transporter (20). Pathway 1 stimulates better
than half of the maltose permease proteolysis observed
following glucose addition but does not contribute to the
rapid inactivation of transport activity. RGT2-1 is a dominant
allele causing constitutive expression of HXT1 (20, 27) and
constitutive proteolysis of maltose permease, that is, degradation even in the absence of glucose (Figure 4) (17).
Pathway 2 uses high rates of glucose transport and phosphorylation to generate an intracellular signal that is primarily
responsible for the glucose-induced inactivation of maltose
transport activity but also stimulates proteolysis (17, 28).
Figure 4 demonstrates that proteolysis is entirely blocked in
the ∆PEST mutant permease, clearly indicating that this
region is the target of both pathway 1 and pathway 2 for
stimulating glucose-induced proteolysis.
Figures 2 and 4 show that N-terminal residues 49-78 are
also required for the rapid glucose-induced loss of maltose
transport activity. In strains expressing the ∆49-78 mutation
or the dileucine motif mutation, the addition of glucose
increases rather than decreases maltose transport activity.
Kinetic analysis of the increased transport activity indicates
that it results from an increase in Vmax with no change in
Km. The simplest explanation is that the mutant permease
proteins are able to continue through the secretory pathway
and localize to the plasma membrane while the wild-type
protein cannot. Glucose appears to block plasma membrane
localization of wild-type permease or redirects it to a different
subcellular site. This glucose-induced altered localization
requires residues 49-78, or possibly just the dileucine motif.
Consistent with this, Hu et al. (29) report a posttranscriptional
block in maltose permease synthesis in the presence of
glucose. These findings are reminiscent of the results that
report differences in the sorting of Gap1p, the general amino
acid permease, depending upon the nitrogen source (30).
The transient increase in maltose transport activity observed following glucose addition to strains expressing the
∆49-78 mutation or the dileucine motif mutation is also
seen in cells expressing wild-type permease but carrying
RGT2-1 (Figure 4) (17) or glucose-treated cells containing
a deletion of REG1 (31). In both of these strains, pathway 1
is active but pathway 2 is not. Thus, it appears that this
increased transport activity is dependent on an activated
pathway 1 but is observed only when pathway 2 is inactive
or when the PEST region is deleted. Loss of the PEST region
substantially reduces the glucose-induced rapid loss in
transport activity in a strain expressing only pathway 2
(Figure 4). This suggests that residues 49-78 are the target
of pathway 2 and, when acted upon or modified in the
presence of glucose, allow maltose permease to be redirected
away from the plasma membrane.
Reg1p is an essential component of pathway 2 (31). Reg1p
is a targeting subunit of protein phosphatase type 1 and binds
to the catalytic subunit Glc7p in the presence of glucose
(reviewed in refs 32-34). A reg1∆ strain exhibits reduced
glucose-induced proteolysis of maltose permease and lacks
the rapid glucose-stimulated loss of maltose transport activity
Biochemistry, Vol. 39, No. 15, 2000 4525
(31). Previous studies demonstrated that Mal61/HA protein
is phosphorylated in maltose-grown cells and that glucose
stimulates increased levels of phosphorylated protein (3). On
the basis of our observation that the relative level of maltose
permease phosphorylation is reduced in reg1∆ strain (31),
we suggest that Reg1p-Glc7p phosphatase indirectly stimulates the phosphorylation of PEST residues, via activation
of a kinase intermediate. Evidence indicates that plasma
membrane localized casein kinase I, encoded by YCK1 and
YCK2, is responsible for the phosphorylation of the PESTlike sequence in uracil permease, stimulating ubiquitination
and degradation of that permease (11, 35, 36). Casein kinase
I is also reported to phosphorylate the SINDAKISS sequence
of Ste2p (7). Preliminary experiments suggest a role for
glucose-activated casein kinase I (37) in the glucose-induced
inactivation of maltose permease (S. Liu and C. A. Michels,
unpublished results).
ACKNOWLEDGMENT
We thank Sabira Ozcan and Mark Johnston for providing
plasmids and disruption primers and Sara Danzi for technical
assistance with in vitro mutagenesis.
REFERENCES
1. Charron, M. J., Dubin, R. A., and Michels, C. A. (1986) Mol.
Cell Biol. 6, 3891-3899.
2. Cheng, Q., and Michels, C. A. (1989) Genetics 123, 477484.
3. Medintz, I., Jiang, H., Han, E. K., Cui, W., and Michels, C.
A. (1996) J. Bacteriol. 178, 2245-2254.
4. Hicke, L. (1997) FASEB J. 11, 1215-1226.
5. Medintz, I., Jiang, H., and Michels, C. A. (1998) J. Biol. Chem.
273, 34454-34462.
6. Hicke, L., and Riezman, H. (1996) Cell 84, 277-287.
7. Hicke, L., Zanolari, B., and Riezman, H. (1998) J. Cell Biol.
141, 349-358.
8. Berkower, C., Taglicht, D., and Michaelis, S. (1996) J. Biol.
Chem. 271, 22983-22989.
9. Kolling, R., and Losko, S. (1997) EMBO J. 16, 2251-2261.
10. Rechsteiner, M., and Rogers, S. W. (1996) Trends Biochem.
Sci. 21, 267-271.
11. Marchal, C., Haguenauer-Tsapis, R., and Urban-Grimal, D.
(1998) Mol. Cell Biol. 18, 314-321.
12. Springael, J. Y., and Andre, B. (1998) Mol. Biol. Cell 9, 12531263.
13. Tan, P. K., Howard, J. P., and Payne, G. S. (1996) J. Cell
Biol. 135, 1789-1800.
14. Brondijk, T. H., van der Rest, M. E., Pluim, D., de Vries, Y.,
Stingl, K., Poolman, B., and Konings, W. N. (1998) J. Biol.
Chem. 273, 15352-15357.
15. Guldener, U., Heck, S., Fielder, T., Beinhauer, J., and
Hegemann, J. H. (1996) Nucleic Acids Res. 24, 2519-2524.
16. Sikorski, R. S., and Hieter, P. (1989) Genetics 122, 19-27.
17. Jiang, H., Medintz, I., and Michels, C. A. (1997) Mol. Biol.
Cell 8, 1293-1304.
18. Elledge, S. J., and Davis, R. W. (1988) Gene 70, 303-312.
19. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. E.,
Seidman, J. G., Smith, J. A., and Struhl, K. (1998) Current
Protocols in Molecular Biology, John Wiley and Sons Inc.,
New York.
20. Ozcan, S., Dover, J., Rosenwald, A. G., Wolfl, S., and
Johnston, M. (1996) Proc. Natl. Acad. Sci. U.S.A. 93, 1242812432.
21. Ozcan, S., Dover, J., and Johnston, M. (1998) EMBO J. 17,
2566-2573.
22. Hochstrasser, M., Ellison, M. J., Chau, V., and Varshavsky,
A. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 4606-4610.
23. Raths, S., Rohrer, J., Crausaz, F., and Riezman, H. (1993) J.
Cell Biol. 120, 55-65.
4526 Biochemistry, Vol. 39, No. 15, 2000
24. Ward, A. C., Castelli, L. A., Macreadie, I. G., and Azad, A.
A. (1994) Yeast 10, 441-449.
25. Gabilondo, A. M., Hegler, J., Krasel, C., Boivin-Jahns, V.,
Hein, L., and Lohse, M. J. (1997) Proc. Natl. Acad. Sci. U.S.A.
94, 12285-12290.
26. Hein, C., and Andre, B. (1997) Mol. Microbiol. 24, 607616.
27. Marshall-Carlson, L., Neigeborn, L., Coons, D., Bisson, L.,
and Carlson, M. (1991) Genetics 128, 505-512.
28. Jiang, H., Medintz, I., Zhang, B., and Michels, C. A. (2000)
J. Bacteriol. 182, 54-64.
29. Hu, Z., Yue, Y., Jiang, H., Zhang, B., Sherwood, P. W., and
Michels, C. A. (2000) Genetics 154, 121-132.
30. Roberg, K. J., Rowley, N., and Kaiser, C. A. (1997) J. Cell
Biol. 137, 1469-1482.
Medintz et al.
31. Jiang, H., Tatchell, K., Liu, S., and Michels, C. A. (2000) Mol.
Gen. Genet. (in press).
32. Johnston, M. (1999) Trends Genet. 15, 29-33.
33. Tu, J., and Carlson, M. (1995) EMBO J. 14, 5939-5946.
34. Tu, J., Song, W., and Carlson, M. (1996) Mol. Cell Biol. 16,
4199-4206.
35. Robinson, L. C., Menold, M. M., Garrett, S., and Culbertson,
M. R. (1993) Mol. Cell Biol. 13, 2870-2881.
36. Vancura, A., Sessler, A., Leichus, B., and Kuret, J. (1994) J.
Biol. Chem. 269, 19271-19278.
37. Estrada, E., Agostinis, P., Vandenheeds, J. R., Goris, J.,
Merlevede, W., Francois, J., Goffeau, A., and Ghislain, M.
(1996) J. Biol. Chem. 271, 32064-32072.
BI992455A